1cs426-winter-2008 Notes  If you’re interested, read Bridson, Hourihan, Nordenstam, “Curl noise for procedural fluid flow”, SIGGRAPH ‘07

2cs426-winter-2008 Time integration woes  For rotational motion, immediately find serious flaw with Forward Euler: instability  Better off using a more accurate, more stable Runge-Kutta method For example: 3rd order

3cs426-winter-2008 Second order motion  Real particles move due to forces Newton’s law F=ma Need to specify force F (gravity, collisions, …) Divide by particle mass to get acceleration a Update velocity v by acceleration Update position x by velocity

4cs426-winter-2008 Basic rendering  Draw a dot for each particle  But what do you do with several particles per pixel? Add: models each point emitting (but not absorbing) light -- good for sparks, fire, … More generally, compute depth order, do alpha- compositing (and worry about shadows etc.) Can fit into Reyes very easily  Anti-aliasing Blur edges of particle, make sure blurred to cover at least a pixel  Particle with radius: kernel function

5cs426-winter-2008 Motion blur  One case where you can actually do exact solution instead of sampling  Really easy for simple particles Instead of a dot, draw a line (from old position to new position - the shutter time) May involve decrease in alpha More accurately, draw a spline curve May need to take into account radius as well…

6cs426-winter-2008 More detailed particle rendering  Stick a texture (or even a little movie) on each particle: “sprites” or “billboards” E.g. a noise function E.g. a video of real flames  Draw a little object for each particle Need to keep track of orientation as well, unless spherical We’ll get into full-fledged rigid bodies later  Draw between particles curve (hair), surface (cloth)  Implicit surface wrapped around virtual particles (e.g. water)